Journal of Criminal Law and Criminology

Volume 49 | Issue 2 Article 4

1958

Research Interviewing in Prison

Donald J. Newman

Follow this and additional works at: https://scholarlycommons.law.northwestern.edu/jclc

Part of the Criminal Law Commons, Criminology Commons, and the Criminology and Criminal
Justice Commons

Recommended Citation
Donald J. Newman, Research Interviewing in Prison, 49 J. Crim. L. Criminology & Police Sci. 127 (1958-1959)

This Article is brought to you for free and open access by Northwestern University School of Law Scholarly Commons. It has been accepted for
inclusion in Journal of Criminal Law and Criminology by an authorized editor of Northwestern University School of Law Scholarly Commons.
 RESEARCH INTERVIEWING IN PRISON

DONALD J. NEWMAN

The author is on the staff of the Department of Sociology in St. Lawrence University, Canton,
New York. During the last three years he has been lecturer on Crime and Delinquency in the Frederick
A. Moran Memorial Institute on Crime and Delinquency which is held annually in his University.
He has done interviewing research in several prisons in Wisconsin, New York and Canada.-EDIToR.

Tom NECESSITY OF IN-PRISON RESEARCH
The necessity for intensive and productive
research in the fields of criminal behavior, juvenile
delinquency and correctional administration is self
evident. Progress in any field, whether physical or
social, ceases when research is meagre or when it is
conducted in a superficial manner. Who could
imagine the present status of medical science with-
out the intensive physiological and biochemical
research on which this profession rests?
For a number of reasons, some custodial, others
bureaucratic, criminological research involving
in-prison populations is neither as frequent nor as
intense as is desirable to provide needed data on the
complex problem of criminal behavior. Crimino-
logical research involves many approaches and
many techniques; a good deal of data can be
gathered in the community and from the courts.
Records, police as well as correctional and psycho-
logical tests, can be examined and analyzed.
Nevertheless, a necessary adjunct to other research
approaches involves interviews with known crim-
inals held in custody in our prisons. Many factors
of sociological significance, such as self conception,
levels of aspiration, attitudes and motives of
offenders, can be measured only by carefully
structured in-prison interviewing. While it is true
that prison populations are not representative
samples of the universe of crime, research among in-
mates is, at the very least, one primary source of
basic criminological data.
THE- NECESSITY OF PERSONNEL COOPERATION
The success of any research conducted within
walls depends very basically upon the mutual
cooperation of the researcher and each of the cor-
rectional personnel, from commissioner tn guard,
with whom the researcher is working. First of all,
permission to conduct the research must come
from the Board or Commissioner of Corrections
and must be supported by the wardens and their
121
assistants. At the same time, it must be made clear
to everyone involved that the Division of Correc-
tions or the Department of Social Welfare is not
sponsoring the research. Prison research sponsored
by correctional authorities or conducted by cor-
rectional personnel suffers at the outset from a
bias which is difficult to overcome. The orientation
and the possible results are too intimately related
to the employment of such individuals to be com-
pletely free from scientific criticism. This does
not mean that correctional personnel are incapable
of research or that they might deliberately bias
their results. Rather, the responses of inmates to
interviewers who stand in an authoritarian posi-
tion to them are open to serious questions of re-
liability and validity.
The most valid research can be conducted by a
trained outsider who has no axe to grind on any
aspect of correctional structure. This in itself,
ho ever, raises a number of problems in gaining
entrance to work behind the walls. There has
been some reluctance on the part of wardens and
other personnel to allow outsiders to enter the
prison to meet in relative isolation with inmate
respondents. Reasons given for this reluctance
generally fall into three general categories: first,
it is argued that the chief function of a prison is
custody and that the entry of an outsider is a
potential threat to the prison's safety. Perhaps the
research might "disturb" the inmates or perhaps
calculating prisoners might use the researcher or
the research situation as a method of escape.
This argument, of course, has some validity.
However, it must be pointed out that no known
instances of prison riots or prison breaks can be
found which can be attributed to research situa-
tions and yet in-prison research by outsiders has
been going on since the time of Lombroso. It is
important, nevertheless, for the warden and others
to make certain that the researcher is well quali-
fied and understands fully the custodial situation
 DONALD J. NREWMAN
and his position in it. Certainly no director of a
hospital would allow an unqualified person to con-
duct medical research but at the same time, few
hospital directors who are dedicated to the al-
leviation of disease would refuse to cooperate with
qualified research personnel.
A second argument relating to this reluctance to
allow outside researchers to enter the gates has to
do with disturbing prison routine. Under present
conditions, all prison personnel certainly have
their hands full pursuing daily duties and cannot
be expected to halt work while research is con-
ducted. The researcher should keep this in mind
in designing his study so that a minimum of dis-
turbance results. Surely a well designed project
will take this into account. A great deal of pre-
liminary research can be done outside the walls and
from prison records. The interviewing itself re-
quires no more than small office space and possibly
the assistance of one guard. Inmates are inter-
viewed individually, under ordinary circumstances
and, with careful planning, no more than one
man per shop should be absent from his work at
any one time. The inmate merely reports for the
interview and following it he returns to his job.
The whole interviewing process should cause less
disturbance in prison routine than a daily sick
call.
A third argument against such research which is
sometimes given is that the research does not
benefit the prison in any way. The study may have
no conceivable relevance to improving opera-
tional procedure or to any other immediate need
the warden may have. Much prison research is, or
can be, basic and theoretical. The goal of crim-
inological research is not only to discover immedi-
ate or "practical" information but to build a solid
foundation of data about criminal behavior on
which the whole science of criminology, and ul-
timately the profession of penology, rests. Again
the analogy between criminological research and
medical research can be drawn. Many, if not most,
important medical discoveries originated from
what was, at the time, impractical and even far-
fetched research.
Underlying the objections of correctional per-
sonnel to outside research there sometimes ap-
pears a fear, even if never expressed, that the re-
search or possible publicity following it might
hurt the prison or jeopardize the jobs of various
prison employees. It should be remembered by
both correctional employees and by the investi-
gators themselves that criminological research is
not concerned with exposing inadequacies in the
[Vol. 49
administration of the prison. This does not mean
that the scientific researcher should, if it is in the
nature of his project, disregard or suppress any
data which might be negative to prison practices.
The researcher is dedicated to objectivity and to
maintaining confidences of respondents and co-
operating authorities. He must report his data as
accurately and as objectively as possible, at the
same time maintaining his confidences and refusing
to single out personel for censure or criticism.
However, the great bulk of criminological re-
search is totally unconcerned with the administra-
tion of any single prison. The prison and its per-
sonnel enter the research picture only insofar as
they are keepers of the inmates whom the re-
searcher desires to study and that their cooperation
is necessary for successful and accurate research.
CONDITrONS FOR SuccEssruL RESEARcH
INTERVIEWING IN PRISON
The ultimate success of any research project
depends upon the accuracy with which data are
gathered. Elaborate analyses or well-written re-
ports are irrelevant unless they are based upon
reliable and valid information gathered from the
research source. Any competent researcher is very
careful that the instruments he uses, such as his
questionnaires, scales, or schedules, are as precise
as.possible. He thinks through his objectives care-
fully; with equal care he builds his hypotheses and
constructs his interviewing forms. He pretests and
possibly pre-codes. However, all of his work is in
vain unless he is equally careful in conducting the
interviews he needs or in administering his tests.
THE SETTING OF THE INTERVIEW
In-prison interviewing presents a number of
problems not encountered in other types of re-
search interviewing. In all sociological, psycho-
logical, and case-work interviewing the actual set-
ting of the interview is considered of primary
importance in building rapport and in obtaining
reliable responses. Attention to the selection of the
setting is particularly important within a prison.
In order to dearly emphasize the distinct nature
of the research project from all things custodial,
the location of the interview should be in a "neu-
tral" or favorable part of the prison.' Preferably
the interviewing should take place in an empty
I See LEwIs ANrnoN DExTER, Role Relationships
and Conceptions of Neutrality in Interiewing. AmaER.
JOUR. OF SocIoL., Vol. LXII, No. 2, Sep., 1956, pp
153-157 and ELEANOR AND NATHAN MACCOBY, The
Interview: A Tool of Social Science in GARDNER LIND-
zEy (ed.), HANDBOOK OF SOCIAL PSYCHOLOGY Cam-
bridge, Mass.: Addison-Wesley, 1954)
 INTERVIEWING IN PRISON
office in the prison school, library, doctor's office
or guidance center. Certainly the use of any cus-
todial employee's office should be avoided. Like-
wise, where possible, the interview should not
be conducted in the psychiatrist's or psychologist's
offices in order to dearly distinguish the research
situation from the more familiar psychiatric
interviews.
The interview setting should be as comfortable
as possible and preferably in a quiet and relatively
isolated office. The interview should be business-
like, and free from any interruptions. The re-
spondent and interviewer should be the only per-
sons in the room during the interview although it
may be necessary to have a guard stationed outside
the door and out of earshot. The guard serves not
only as a security measure but may appreciably
aid the interviewing situation by sending for
respondents when they are needed and maintain-
ing order should an inmate have to wait for a few
minutes until the interviewer can see him.
BUILDING RAPPORT IN PRISON INTERVIEWS
The question of rapport development in research
interviews is ordinarily somewhat different than
in case-work or psychiatric interviewing which
may involve a long series of interviews with the
•same respondent. In sociological research, one or
two interviews with a respondent is sometimes
sufficient, although the actual number of inter-
views and lengths of each depend upon the nature
of the project. This means that to obtain reliable
responses, rapport of adequate intensity must be
developed rapidly. Obviously, mutual trust and
confidence is somewhat more difficult to come by
when the respondent is incarcerated and when he is
seemingly arbitrarily summoned for an interview
that he knows little or nothing about. Conse-
quently, the researcher must be prepared to suc-
cintly familiarize the inmate with the -purposes of
the research in terms that he can comprehend and
to seek his voluntary assistance. It is not neces-
sary, of course, to explain fully the hypotheses
underlying the research and it is both unscientific
and unwise to tell the inmate precisely the data
sought.2 Yet enough information about the project
must be explained in order to prevent miscon-
ceptions and damaging yard rumors.
2 CaassEY, for example, successfully concealed the
theory underlying his research with embezzlers.
DONALD R. CRESSEY, OTHER PEOPLES MONEY: A
STUDY ON THE SOCIAL PSYCHOLOGY OF EMBEZZLE-
mENT, Glencoe, Illinois: The Free Press, 1953, p. 25.
See also ALFRED LINDESMITH, OPIATE ADDICTION,
(Bloomington, Indiana: Principia Press, 1947) p 6.
Even before, or certainly while explaining the
research project, the researcher should dearly
identify himself and emphasize the fact that he is
totally unrelated to the prison administration and
that any information divulged by the inmate will
be held in strict confidence and will not be dis-
cussed with any of the prison staff nor for that
matter with anyone else. It is perhaps a good idea
for those researchers who have university affilia-
tion to carry with them a staff directory or univer-
sity bulletin so that any doubting inmates may
look up their name and position. While this does
not really prove identity, it has been found helpful
by this writer and seems to be sufficient evidence
for most inmates. As Dexter has said-"The role
assigned to the interviewer by the informant will
be in terms of the informant's conception of the
interviewer's group affiliations." 3
Vital to rapport development and the ultimate
reliability of the data obtained is frankness con-
cerning the relationship of the interview to the
inmate's particular case. It must be strongly em-
phasized that the interview and the total results
of the research project will have no effect in miti-
gating the inmate's sentence, in increasing his
chance for parole, or in any other way acting as a
direct benefit to him. Conversely, the interview or
project will have no harmful effect on his record.
If, under such conditions, he sees no incentive for
cooperation, long-range effects of such research
may.be pointed out to him as well as illustrations of
general progress in criminology and penology re-
sulting from just such research as the present
project.
The information sought in the interview should
be elicited as naturally as possible to build and
maintain rapport throughout the interviewing
period. Any questions should be frankly answered
and the respondent should be freely allowed to see
what notes have been taken. Again, in recording
notes during the interview, the confidential nature
of the project should be stressed. The problems of
recording in-prison interviews are approximately
the same as in other types of research. Some in-
mates show reluctance to have their statements
written, possibly because of negative police and
court experience, but others are only too eager to
speak "for the record." Cressey has used a variety
4
of recording techniques with evident success.
Notes, schedules, and other forms should be kept
securely filed by the researcher at all times and
3
DEXTER, op. cit. p 153.
4 CPESSEY, op. cit. pp 24-26.
 DONALD J. NEWMAN
no final typing or coding should be done by
prison personnel.
The entire deportment of the researcher should
be above reproach during the extent of the total
research project. He must bear in mind that he is
seeking to obtain reliable information from a more
or less initially hostile population. To gain and
maintain inmates' confidence and to prevent harm-
ful grapevine rumors that might destroy his
work, he should remain as aloof as possible from
prison personnel even during the off-hours of
lunch and evenings.
SPECIAL PROBLEMS OF INTERVIEWING ORDER
The skilled researcher should give some thought
to the order in which he interviews his sample. If
possible, he should check with prison personnel
concerning the reputation of the first few respond-
ents. He should avoid beginning his project by
summoning unpopular or sexually deviant in-
mates in order to prevent his study from being
negatively defined in the yard before it has a
chance to begin operating. He should try to start
with "average" inmates in terms of type of
offense for which sentenced, length of sentence
and so on. This does not mean that all his respond-
ents should be these "average" inmates; he is
bound to interview his sample without such arbi-
trary substitution. However, he should keep in
mind what is probably the most unique char-
acteristic of prison research, namely the fact that
most or all members of his sample know one
another, share with one another certain senti-
ments, orientations and experiences of prison
existence and will undoubtedly communicate with
one another and quite probably will collectively
evaluate the research project before it has really
begun. 5 It is, therefore, important that the first
few respondents carry back to the yard a straight
story of the research project and it is equally im-
portant that their in-prison reputations are such
that the project can be evaluated accurately by
other inmates.
If the research is concerned with gathering facts
about the crimes for which the inmates have been
sentenced, it would seem advisable to pay some
attention to the interviewing order of men sen-
tenced as partners in their crime. If, by careful
scheduling, one partner is prevented from com-
municating with the other partner following the
interview, a crude measure of reliability can be had.
I For a similar situation see MAURICE LEZNOFF,
Inler'iewing Homosexuals AMER. JoUR. OF SOCIOL.,
Vol. LXII, No. 2, Sep., 1956, pp 202-204.
[Vol. 49
SPECIAL PROBLEMS OF ARGOT
Prison slang and other variations in criminal
argot inevitably present a problem to the in-
prison researcher. Obviously, the interviewer must
clearly and accurately understand the responses
of his sample in order to gather valid data and the
peculiar language variations of many inmates
presents a real challenge. In general, argot en-
countered in prison interviewing falls into five
categories: first, slang concerning incarceration,
prison routines and prison personnel and this may
include terms common only to one specific prison;
second, inmate terms for various legal processes
and types of sentence; third, argot common to
certain types of offenses and offenders varying
from terms used by burglars to the language of
confidence men; fourth, argot common to sex
practices and sexual deviations; and last, argot
common to various vices which sometime accom-
pany crime like gambling and narcotic addiction.
These variations, of course, are in addition to
ordinary slang differences created by the usual
amount of social distance which exists between the
interviewer and his respondents. Many inmates
are slum-produced and the ordinary "jive-talk" of
the slum is often unfamiliar in the academic world
of the researcher. The interviewer should familiar-
ize himself as much as possible with these argots
which he may do by asking correctional authori-
ties, reading a number of unofficial glossaries of
various slangs, and, of course, by asking the mean-
ing of terms when he encounters them in the
interview.
It is ordinarily not necessary nor desirable for a
researcher to be able to talk like a confidence man,
yet as anthropological research must take place
in the language of the culture being studied, so the
prison researcher should make an effort to learn at
least some of the basic "criminalese" encountered
in his work. 6 Frequent interruptions to clarify
terms damage rapport and the strained, often
erroneous use of argot by the interviewer may
destroy rapport completely. As Kinsey has
pointed out, there is considerable variation in the
use of vernacular terms by locale, by racial groups,
and between generations,7 and the interviewing
structure must be flexible enough to bridge mis-
understanding. The rigid techniques of question-
6 See, for example, the excellent description of the
role necessarily assumed by the interviewer in DONALD
RASMUSSEN Prisoner Opinions About Parole AMER.
SOCIOL. REV., Vol. 5, No. 4, Aug., 1940, pp 589-590.
7ArYED KINSEY and others, SEXUAL BEIIAVIOR IN
THE HUMAN MALE, New York: Saunders, 1948, p 52.
19581 INTERVIEWING IN PRISON
naires and directed interviews leave something
to be desired in research with prison populations;
accurate understanding can be developed best in
an interview situation. As Caplow has said: "The
expert interviewer is much more than a recording
device. No matter what the form of the interview,
he should pursue his questioning to the point
8
where no significant ambiguities exist for him."
PROBLEMS oF RELI_43ILITY AND VALIDITY IN
PRISON RESEARCH
Fortunately, information from in-prison inter-
views is, in some ways, easier to check for relia-
bility than interviews taken "in the open." Each
inmate has an extensive dossier containing a good
deal of pertinent information about his back-
ground, criminal record and so on. Many times
modern prison records contain statements in the
inmate's own words regarding his past life, the
conditions surrounding his criminal career, his re-
lationship with authorities, his parents and others
who know him. In short, a brief but verbatim
autobiography exists. This is, of course, in addition
to test results, investigator's reports, court records,
psychologists' summaries and the other com-
ponents of penal records. Certainly these are indis-
pensible as a reliability check on interviewing data.
Truthfulness of responses is not a problem
peculiar to prison research, but is an issue in all
projects involving interviewing. Probably because
"imprisonment" and "dishonesty" are popularly
related conceptions, doubts concerning data from
inmates are more often expressed than doubts of
data from, say, public opinion polling. There is, of
course, no absolute way of measuring the degree
of honesty in expressed attitudes, or sincerity, or
the accuracy of recalled emotions. Nevertheless, all
possible methods of validation should be used,
and, of course, care should be taken in the design
of the study and the interviewing techniques to
assure in advance research situations in which
honesty and accuracy can be engendered. Merton,
for example, has claimed that the focused interview
is the best technique for validly measuring sub-
jective experiences such as attitudes and emo-
tional responses9 and Young reports the successful
10
use of an oral interviewing guide.
8THEODORE CAPLOW, The Dynamics of Information
Inter-o'iing AMxER. JOUR. oF SocIOL., Vol. LXII, No.
2, Sep., 1956, p 167.
9ROBERT MERTON, MARJORIE FIsKE and PAxRiCiA
L. KENDALL, THE FocusED INTERvIEw Glencoe: The
Free Press, 1956
10PAULINE V. YOUNG, Scientific Study of Young,
Occasional, Urban Male Offenders AMER. SOcIOL. REV.,
Vo. 5, No. 4, Aug., 1940, p 596-600.
In addition to study design and interviewing
technique, many other methods have been de-
veloped to check validation, such as comparing
results of data with subsequent events, compari-
son of results with other, similar research, com-
paring pre-testing results with final results, validat-
ing data against judgement of experts, repeated
check interviewing and so on." As far as research
with inmates is concerned, if Caplow's hypothesis
that ". . . being interviewed is an inherently satis-
fying experience and ordinarily constitutes its
own goal"' 2 is correct, the components of this hy-
pothesis, escape from boredom, attention, a
chance to express one's views and the other satis-
factions received by the interviewee will be even
stronger motives for free and frank conversation in
inmate interviewing. Coupled with the opportuni-
ties for the researcher to use already compiled
records of many facets of his respondents' lives,
the entire in-prison situation seems to lend itself to
a degree of accuracy and methods of reliability
check not afforded in many other types of research.
THE ETHICAL POSITION OF THE PRISON
RESEARCHER
Successful interviewing necessitates a confi-
dential relationship between interviewer and
respondent of the same intensity as doctor-patient
or lawyer-client relationships. However, the social
researcher-respondent situation is not privileged
in the law to the same extent as are these other
relationships. Nevertheless, to be assured of re-
liable data, research interviewers inevitably inform
their respondents that all information will be held
in confidence but often qualify this assurance by
explaining that data will be used "only in a statis-
tical way" or that all identifying information will
be concealed. Presumably, this assurance of con-
fidence extends to any information gathered in the
interviewing situation even if it is not actively
sought and is irrelevant to the particular research
problem.
In dealing with inmates, the researcher is in a
position where, inadvertently perhaps, he might be
informed of unsolved crimes, of plans to disturb
prison routine or to break jail, of corruption among
guards, or of vice within the walls. The issue he
then faces is whether to ignore the information,
inform prison or police authorities, or to personally
idscourage any inmate who reveals such informa-
" For a discussion of validation techniques see
MILDRED PARTEN, SURVEYS,' POLLS AND SAMPLES,
New York: Harper, 1950, pp 486-498.
p 169.
12 CAPLOW, op. Cil.,
 DONALD J. NEWMAN
tion from actively participating in the illegal or
disruptive activity. The answer, should such a
dilemma occur, cannot be arbitrarily formulated.
Each situation must be judged as it occurs and the
researcher must weigh the factor of his position as
a law-abiding citizen against his scientist's role of
keeping confidences. An understanding warden
once told me, at the start of a research project,
that if I should learn of any dangerous or dis-
ruptive plans among the inmates (this was at the
height of the recent prison riot fad), I should
keep this information to myself. He explained that
his job was custody, mine was not, and that he felt
perfectly competent to maintain custody with his
own staff.
Should an issue arise which involves the possible
loss of human life or the commission of another
felony, the researcher must decide as his own
conscience and the law to direct. However, any
prison community abounds in unfounded rumors
and it is not too unlikely that some inmates might
create a fictional story of an impending break or
other violation to test the promised confidence of
the researcher. The skilled prison interviewer
should act very cautiously on such information. If
men are really planning a break or a riot it seems
unlikely that an interviewer would be told, cer-
tainly not at the beginning of the study before any
rapport has a chance to develop. Reporting such
fictional information to prison authorities would
undoubtedly seriously damage the research project.
Ultimately, however, the researcher must make
each decision as the case arises and while he does
not have the immunity of the confessional, at the
same time he has certain obligations to his own
scientific study, and one of these is keeping infor-
mation confidential within all reason.
CRIMINOLOGY AND IN-PRISON INTERVIEWNG
Criminology, as a special field within the gen-
eral area of social science, is concerned with many
[Vol. 49
types of behavior and variations in law-breaking
in addition to those represented in prison samples.
Techniques in criminological research include de-
vices other than the interview. Nevertheless, face-
to-face sources of data are vital to progress in
any science of human behavior. Donald Taft has
said that "The hope of ultimate control and pre-
vention of crime depends upon the future of
criminological research"' 1 and research with prison
inmates can provide, much more than it has,
valuable additions to knowledge about law viola-
tions. Not all answers can be obtained by inter-
viewing inmates; the study of conduct norms, or
white-collar crimes, statistical analyses of crime
rates, those rare studies of the ordinary criminal
"in the open," and other such research contributes
to the central body of knowledge. However, in-
prison interviewing has been chiefly information
oriented, demographic in content. As Sykes has
so aptly stated:
"In the past criminology has been often a borrower
but seldom a lender, as far as a theory of human be-
havior is concerned. It has drawn heavily on the dis-
ciplines of psychology, sociology, jurisprudence, and
so on, in order to solve its questions. It has rarely con-
tributed original propositions or substantiating em-
pirical studies to the social sciences. As a result the
criminologist has become somewhat isolated from the
main body of social theory and research; and the study
of crime, in spite of the inherent interest of the subject
matter, is in danger of becoming a dull parade of un-
attached facts."' 4
Carefully constructed research projects and
skillfully conducted in-prison interviewing can do
much to lessen this danger.
11DONALD R. TAFT, CRIMINOLOGY, New York, New
York: The Macmillan Company, 1956, p 20.
" G.EsnAm M. SYrcEs, CRam An SocrTY, New
York, New York: Random House, 1956, pp 115-116.